04Ecommerce / Retail
A shopping copilot on cleaned catalog embeddings — raising add-to-cart by 12%.
Timeline · 7 weeks
01Challenge
Shoppers couldn’t find products because search didn’t understand attributes or intent. Catalog data was inconsistent — missing fields, duplicate listings, messy naming.
Merchandising wanted better discovery without a multi-year PIM rewrite.
02Approach
We cleaned and embedded the catalog so attributes like size, material and use-case became searchable signals — not buried text.
A shopping copilot sits on top of that retrieval layer, helping shoppers describe what they want in natural language.
Attribute search and recommendations were wired into the storefront experience the brand already operated.
03Results
Add-to-cart rate rose 12% after launch.
Catalog quality improved enough that search and filters became trustworthy again.
The brand owns the pipeline, embeddings workflow and storefront integration end to end.
04Project details
Deliverables
- Data cleaning & embeddings
- Shopping copilot
- Attribute search
Stack
- OpenAI
- Pinecone
- Shopify
- Next.js
- Industry
- Ecommerce / Retail
- Timeline
- 7 weeks
06Start a project
Tell us what you're building, where you are today, and what needs to happen next. We'll come back with a technical point of view — not a sales pitch.
- Response within one business day
- Scoping workshop available within the week
- Worldwide, remote-first